Machine Learning for Early Diagnosis of Endometriosis(MLEndo)

Recruiting

• Conditions: Infertility, Female; Endometriosis; Pelvic Pain
• Interventions: Diagnostic Test: Self reported data collection

• Registration Number: NCT06147687
• Lead Sponsor: Semmelweis University

Brief Summary

The project aims to create a large prospective data bank using the Lucy medical mobile application and collect and analyze patient profiles and structured clinical data with artificial intelligence. In addition, authors will investigate the association of removed or restricted dietary components with quality of life, pain, and central sensitization.

Detailed Description

Introduction: Endometriosis is a complex and chronic disease that affects ∼176 million women of reproductive age and remains largely unresolved. It is defined by the presence of endometrium-like tissue outside the uterus and is commonly associated with chronic pelvic pain, infertility, and decreased quality of life. Despite numerous proposed screening and triage methods such as biomarkers, genomic analysis, imaging techniques, and questionnaires to replace invasive diagnostic laparoscopy, none have been widely adopted in clinical practice.

. Despite the availability of various screening methods (e.g., biomarkers, genomic analysis, imaging techniques) that are intended to replace the need for invasive diagnostic laparoscopy, the time to diagnosis remains in the range of 4 to 11 years. Aims: The project aims to create a large prospective data bank using the Lucy medical mobile application and collect and analyze patient profiles and structured clinical data with artificial intelligence. In addition, authors will investigate the association of removed or restricted dietary components with quality of life, pain, and central sensitization. Methods: A Baseline and Longitudinal Questionnaire in the Lucy app collects self-reported information on symptoms related to endometriosis, socio-demographics, mental and physical health, nutritional, and other lifestyle factors. 5,000 women with endometriosis and 5,000 women in a control group will be enrolled and followed up for one year. With this information, any connections between symptoms and endometriosis will be analyzed with machine learning. Conclusions: Authors can develop a phenotypic description of women with endometriosis by linking the collected data with existing registry-based information on endometriosis diagnosis, healthcare utilization, and big data approach. This may help to achieve earlier detection of endometriosis with pelvic pain and significantly reduce the current diagnostic delay. Additionally, authors can identify nutritional components that may worsen the quality of life and pain in women with endometriosis; thus, authors can create evidence-based dietary recommendations.

Keywords: Endometriosis, Machine learning, Non-invasive diagnosis, Diet

Study Timeline

• Study Start: 2022-01-01
• Status Verified: 2023-09
• Study First Submitted: 2023-09-25
• Study First Submitted QC: 2023-11-21
• Last Update Submitted: 2023-11-21
• Study First Posted: 2023-11-28
• Last Update Posted: 2023-11-28
• Primary Completion: 2024-12-31
• Study Completion: 2024-12-31

Recruitment & Eligibility

• Status: RECRUITING
• Sex: Female
• Target Recruitment: 10000

Inclusion Criteria

• Women in reproductive age
• 5000 patients with endometriosis
• 5000 patients without endometriosis

Exclusion Criteria

• Ongoing pregnancy
• Malignant condition of ovary/uterus/breast

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Arm && Interventions

Group	Intervention	Description
Control	Self reported data collection	5 000 people in a control group will be enrolled and followed up for 1one year. To participate in the study, the women must meet the inclusion criteria.
Patients with endometriosis and Healthy controls	Self reported data collection	5 000 people with endometriosis will be enrolled and followed up for 1one year. To participate in the study, the women must meet the inclusion criteria.

Primary Outcome Measures

Name	Time	Method
Patient- profiling using the Lucy app	24 month	Establish a comprehensive and extensive prospective big data repository using the Lucy app. This initiative aims to identify unique clinical cohorts by leveraging various factors such as digital footprints, symptoms, patient experiences, comorbidities, clinical severity, and lifestyle patterns. By employing Using ML for big data analysis, authors can build patient profiles and structured clinical data that facilitate the early detection of endometriosis with pelvic pain.; Self-reported data of the participants will be measured as follows: * Evaluating the quality of life using the 5-level EQ-5D (EQ-5D-5L); * Endometriosis Health Profile 5 (EHP-5) .; * Pain scores using the Visual Analogue Scale (VAS) .; * Central pain sensitization using the short version of Central Sensitization Inventory (CSI-9)

Secondary Outcome Measures

Name	Time	Method
Impact of diet and lifestyle on the development of endometriosis	24 month	Additionally, authors can identify nutritional components that may worsen the quality of life and pain in women with endometriosis; thus, they can create evidence-based dietary recommendations.; The changes in quality of life will be assessed by using Self-reported data of the participants will be measured as follows: Change From Baseline in Pain Scores on the Visual Analog Scale at 12 months. Changes from baseline values on EHP5 at 12 months

Trial Locations

Locations (2)

Bokor Attila; 🇭🇺 Budapest, Hungary

Semmelweis University; 🇭🇺 Budapest, Hungary